Introduction
The past decade has witnessed a growing interest in the study of geography and economics, on the one hand, and political values, attitudes, and behavior, on the other. Concepts such as “rural resentment” (Cramer, Reference Cramer2016), “the revenge of the places that don’t matter” (Rodríguez-Pose, Reference Rodríguez-Pose2018), and “place-based resentment” (Munis, Reference Munis2022) have been coined to gauge that individuals residing in rural places use the ballot to strike back at urban liberal elites. The empirical link between these sentiments and the rise of what has been termed the “populist backlash” has been a focal point of research, notably evident in the strong support for such movements in certain geographies. This development led Rachman (Reference Rachman2018) to state that urban-rural splits have become “the great global divide” in contemporary politics.
Moreover, the influential work by Cramer (Reference Cramer2016) emphasizes the importance of place identity in how people understand and interpret politics. Cramer argues that rural Americans develop their preferences based on a unique sense of identity tied to their place of residence and a feeling of being disadvantaged compared to the influential urban elites. This resentment toward urban elites could lead rural residents to reject policies proposed by urban politicians while endorsing equivalent policies put forward by rural politicians.
In line with these assumptions, previous research suggests that the actors behind a policy issue are crucial in shaping public opinion (Nicholson, Reference Nicholson2012). However, such studies predominantly focus on party cues. Moreover, recent research suggests that place cues can influence preferences. However, these studies have primarily focused on outcomes such as evaluations of political candidates (Haffert et al., Reference Haffert, Palmtag and Schraff2023; Jacobs & Munis, Reference Jacobs and Munis2019) and the allocation of government resources (Lyons & Utych, 2021). For instance, Jacobs and Munis (Reference Jacobs and Munis2019) found that rural respondents viewed urban-associated candidates more negatively than rural-associated candidates and that place-based cues were less relevant for urban residents. Thus, while no study has explored how place-based cues affect policy agreement, these studies lend some suggestive support in line with this idea.
Although studies in other contexts have begun to surface – that is, Huijsmans et al. (Reference Huijsmans, Harteveld, van der Brug and Lancee2021), De Lange et al. (Reference De Lange, van der Brug and Harteveld2023), Stein et al. (Reference Stein, Buck and Bjørnå2021), Kenny and Luca (Reference Kenny and Luca2021), Haffert (Reference Haffert2022), and Huijsmans and Rodden (Reference Huijsmans and Rodden2025) – the bulk of empirical evidence on urban-rural polarization stems from Anglo-Saxon contexts (i.e., the USA, UK, and Canada, cf. Huijsmans and Rodden, Reference Huijsmans and Rodden2025). These countries employ majoritarian, first-past-the-post electoral systems, which are known for producing fewer competitive political parties. An argument that has been proposed is that the fewer the parties, the more geographical polarization will be observed. As American journalist Ezra Klein (Reference Klein2020) highlights, in systems where a few parties dominate, politics tends to become binary and polarized due to intense competition and ideological divergence (cf. Rodden, Reference Rodden2019).
However, if urban-rural splits have become that “great global divide” in contemporary politics Rachman (Reference Rachman2018) speaks of, they should also be present in egalitarian multi-party systems. Indeed, a handful of recent studies conducted in Europe suggest that rural resentment extends beyond the USA, UK, and Canada (see, e.g., De Lange et al., Reference De Lange, van der Brug and Harteveld2023; Hegewald, Reference Hegewald2023; Huijsmans, Reference Huijsmans2023; Zumbrunn, Reference Zumbrunn2024a, Reference Zumbrunn2024b). That said, we believe it is crucial to further extend this research to better understand the existence and consequences of rural resentment in countries that traditionally have been less politically polarized and have more extensive territorial cohesion ambitions and redistribution schemes across regions and municipalities.
Moreover, Cramer’s (Reference Cramer2016) work has convincingly demonstrated that rural dwellers tend to believe the government neglects rural areas when distributing resources and that politicians are unconcerned with the problems afflicting their geographies (cf. van Vulpen, Reference Van Vulpen2025). Hence, rural resentment looms large in her work. However, this might be because all informants are rural dwellers. It is far from implausible that urbanites also identify with their place and that such place identification might translate into politicized resentments vis-á-vis out-groups (e.g., Munis, Reference Munis2022).
In other words, just as rural citizens may feel disadvantaged relative to urbanites by government programs, some urban citizens may resent electoral rules that favor rural areas and diminish their influence or feel that the government wastes money in rural areas. On the other hand, people in urban areas may well agree with rural citizens that rurality has been mistreated and that the political system is stacked against such geographies, or that place-identity may not be a meaningful identity in the political world simply because their place identity is not as salient to them (Hegewald & Schraff, Reference Hegewald and Schraff2022; Miller et al., Reference Miller, Gurin, Gurin and Malanchuk1981). Recent research suggests that place-based resentment exists among urban residents as well, although this resentment is asymmetric, with rural residents harboring higher levels of place-resentment compared to their urban counterparts (Borwein & Lucas, Reference Borwein and Lucas2023; Haffert et al., Reference Haffert, Palmtag and Schraff2023; Shea and Jacobs, Reference Shea and Jacobs2023; Lyons & Utych, Reference Lyons and Utych2023; Munis, Reference Munis2022). That said, surprisingly little is known about the extent to which affective polarization and in-group/out-group thinking differ between rural and urban populations.
Against this backdrop, our study aims to expand and deepen the knowledge about place-based resentment and its consequences on geographical polarization. The study’s main contribution is to investigate the causal effect of place identity cues on agreement with political statements. Moreover, we contribute to the literature by (a) turning to a non-Anglo-Saxon context, characterized by being a unitary state, having a multi-party system without a history of the urban-rural divide being politicized, a tradition of consensual decision-making, and recurrently referred to as “egalitarian” – Sweden, and (b) by investigating the occurrence of place-based identities and place-based resentment among both rural and urban residents.
Conceptual framework and hypotheses
Social identity theories explain why individuals categorize themselves and others into distinct in- and out-groups, forming positive emotional attachments to their in-group while engendering negative feelings toward the out-group (Haslam et al., Reference Haslam, Ellemers, Reicher, Reynolds and Schmitt2010; Tajfel & Turner, Reference Tajfel, Turner and Worchel1979). These attachments may fuel resentment and intergroup conflict, particularly in political contexts where competition over scarce resources is a central concern. In this context, we conceptualize place-based affective polarization as an overarching framework that captures how geographic identities give rise to polarized emotional, cognitive, and evaluative responses toward in- and out-groups. This framework encompasses three interrelated dimensions: place-based identity, place-based resentment, and place-based stereotyping.
Place-based identity refers to individuals’ sense of belonging to a community defined by geographic location and a psychological connection to others who share that place. Scholars across disciplines have explored how people develop meaningful connections with the places they live, often rooted in shared experiences, cultural norms, and local narratives (e.g., Agnew, Reference Agnew2014; Cuba & Hummon, Reference Cuba and Hummon1993; Lalli, Reference Lalli1992). Importantly, identity itself can serve as a heuristic in information processing, where people simplify judgments by evaluating information based on its alignment with their group identity (Tinghög, Barrafrem & Västfjäll, Reference Tinghög, Barrafrem and Västfjäll2023).
Place-based resentment refers to the perception that one’s geographic in-group has been unfairly disadvantaged relative to a geographic out-group, accompanied by emotional hostility toward those perceived as benefiting undeservedly. It emerges in the intersection between place identity and geographic grievances, that is, when individuals strongly identify with a particular geographical location and perceive that political or institutional authorities neglect or mistreat their community relative to others (Cramer, Reference Cramer2016; Miller et al., Reference Miller, Gurin, Gurin and Malanchuk1981; Munis, Reference Munis2022). For instance, rural resentment may stem from perceptions that rural areas are politically marginalized and underfunded (e.g., Rickardsson et al., Reference Rickardsson, Mellander and Bjerke2021). Conversely, urban resentment might spring from a perception that rural residents benefit more from government resources without contributing equally in taxes (e.g., Pew Research Center, 2018). In both cases, resentment is rooted in geographic comparisons that fuse identity, affect, and perceived injustice.
Place-based stereotyping refers to the simultaneous operation of in-group favoritism and out-group derogation based on geographic identity (see e.g., Hegewald & Schraff, Reference Hegewald and Schraff2022). This concept draws inspiration from research on affective polarization, which demonstrates that political partisans increasingly express strong emotional attachment to their in-group alongside deep aversion toward opposing parties (e.g., Boxell et al., 2020; Iyengar et al., Reference Iyengar, Lelkes and Levendusky2019; Reiljan, Reference Reiljan2020; Wagner, Reference Wagner2021; Bäck et al., Reference Bäck, Carroll, Renström and Ryan2023). While these studies focus primarily on partisan identities, the logic of affective polarization can be extended to geographic identities, where place becomes a salient social category. In this context, place-based stereotyping refers to how people form rigid and emotionally charged trait evaluations of others based on their geographic affiliation. These evaluations include both positive stereotypes about one’s in-group (such as viewing rural people as authentic and hardworking, or urban people as dynamic and open-minded) and negative stereotypes about out-groups (such as perceiving rural residents as narrow-minded and intolerant, or urban dwellers as self-centered and arrogant). These stereotypes function as social heuristics that may reinforce geographic polarization and inform attitudes toward political actors and policies.
Hypotheses
We set out to test one main hypothesis and three descriptive hypotheses. The descriptive hypotheses explore the existence of rural-urban tension in Sweden. The main hypothesis investigates the causal effect of place-based cues on agreement with political statements (see Hypothesis 4).
For the descriptive hypotheses, we draw on previous literature on place-based polarization observed in other countries. First, previous studies suggest that place-based in-group identity tends to be more salient or meaningful among rural people compared to their urban counterparts (see e.g., Munis, Reference Munis2022). Similarly, we expect a stronger place-based identity among rural people than urban ones (H1). Second, studies conducted in the USA and UK highlight the experiences of rural residents facing socioeconomic decline, expressing a desire for recognition and control over their lifestyles, leading to place resentment (Cramer, Reference Cramer2016; Jennings & Stoker, Reference Jennings and Stoker2016; Walsh, Reference Walsh2012). Therefore, we anticipate that rural citizens in Sweden, who face unique challenges and opportunities (see section 3.1), will exhibit higher levels of place resentment compared to their urban counterparts (H2). Third, because place-based identities and place-based resentment are assumed to be more salient for rural individuals than for urbanites, we expect rural individuals to hold stronger place-based stereotypes. Specifically, rural individuals are expected to show more favorable in-group and more negative out-group evaluations than urban individuals (H3).
For our main hypothesis, we build on the literature on partisan source cues, which has found that people tend to evaluate political information through a lens colored by their identity. Because place identity has been suggested as an increasingly important factor in understanding how citizens interpret political information (Cramer, Reference Cramer2016), individuals may use place cues as a heuristic when evaluating and interpreting political information. Just as previous literature on partisan source cues has found that individuals often evaluate political information through a lens shaped by their social identity (e.g., Cohen, Reference Cohen2003; Ehret et al., Reference Ehret, Van Boven and Sherman2018; Erlandsson et al., Reference Erlandsson, Nilsson, Rosander, Persson and Van Boven2024; Nicholson, Reference Nicholson2012; Tannenbaum et al., Reference Tannenbaum, Fox and Rogers2017; Van Boven et al., Reference Van Boven, Ehret and Sherman2018; Verkuyten & Maliepaard, Reference Verkuyten and Maliepaard2013), we expect that in-group cues based on place-identity will generate more agreement with political statements than out-group cues based on place identity.
Thus, we predict that both rural and urban respondents will evaluate political statements more favorably when presented by a politician from their place-based in-group compared to their place-based out-group (H4). In other words, we test if rural respondents will agree more with political statements from rural politicians (H4a), and if urban respondents will agree less with political statements from rural politicians (H4b).
Case, data, and methods
The research setting: Sweden
Sweden is chosen because it is far from self-evident that polarization between urban and rural dwellers is present. On the one hand, there are four reasons to suspect that Sweden would be a relatively tough test to find polarization: (1) It has historically been recognized as a rational, consensus-oriented, pragmatic political culture (Anton Reference Anton1969; Castles Reference Castles1976); (2) it has had high ambitions to even out geographical differences between municipalities, including an ambitious municipal equalization scheme (Söderdahl 2015); (3) it has had no history of the urban-rural or center-periphery dimension being seriously politicized by political parties (Oscarsson & Holmberg Reference Oscarsson and Holmberg2016; Demker & Svåsand Reference Demker and Svåsand2005; Christensen 2005); (4) it is a multi-party system, and such systems are assumed to ameliorate geographical tensions (Rodden, Reference Rodden2019; Klein, Reference Klein2020).
On the other hand, two developments could potentially have driven rural dissatisfaction and urban-rural polarization: (1) Sweden’s metropolitan areas have grown considerably, while smaller municipalities, especially in the country’s rural and sparsely populated areas, have experienced dramatic depopulation (Mörk et al. Reference Mörk, Erlingsson and Persson2019). With fewer working-age individuals supporting an increasing portion of elderly, nonreproductive, and nonworking individuals, the capacity to deliver high-quality welfare services has weakened (SOU 2020:8); and (2) the state has actively withdrawn from sparsely populated areas, reducing its presence through institutions such as the Police, the Public Employment Service, and the Social Insurance Agency (e.g., Erlingsson et al. Reference Erlingsson, Öhrvall and Wallman Lundåsen2023).
Sweden’s complex landscape makes it difficult to predict how place-based resentment might evolve. Its history of consensus and geographic redistribution contrasts with ongoing urbanization and centralization, making it especially relevant to test whether theories of place-based bias apply in the Swedish context.
Design and procedure
Sample and procedures: To test our hypotheses, we implemented a preregistered online survey experiment with a Swedish sample (n = 2,051), where 52.2% self-identified as rural and 47.8% as urban (46% male; mean age = 52 years; see Appendix A1 for details on sampling procedures). The preregistration is available at https://osf.io/2f9rc. See Appendix F and Table F1 for information about the correspondence between the manuscript and the preregistration document. Before launching the main study, we conducted a pilot study with a convenience sample (Full pilot details are available in Appendix D).
The study design follows a sequential structure (Figure 1), encompassing five stages that operationalize distinct components of our conceptual framework: place-based identity, place-based resentment, and place-based stereotyping. This design enables us to assess both descriptive patterns (e.g., identity strength, resentment, and stereotyping) and whether place-based cues causally affect agreement with political statements, thereby testing for place-based political bias.
Figure 1. Overview of the experimental design and measurement stages.
Place-based bias: In Stage 1, we measure support for policy proposals with three political statements presented in the form of tweets. These statements address prominent policy areas in Sweden – healthcare, education, and immigration – adapted from Bäck et al. (Reference Bäck, Carroll, Renström and Ryan2023) to ensure topical relevance and emotional salience (see Table 1). The selected topics reflect issues that are central to the Swedish political agenda and likely to elicit attitudinal responses across the ideological spectrum. Importantly, the wording of each statement was deliberately crafted to be ideologically neutral and contextually plausible, as if originating from any political party or region. For example, the tweet about immigration stated, “The latest statistics show that the total number of immigrants worldwide was around 82,4 million. Many of them will probably need to seek asylum in Sweden. We need a new politics that takes this seriously.”
To examine the causal effect of geographic source cues on political agreement (i.e., place-based political bias), we employed a between-subjects experimental design. Each respondent was randomly assigned to view one of the three political statements, attributed either to a politician based in Hammerdal (rural cue) or Stockholm (urban cue). The geographic cue is embedded in a textual frame introducing the tweet, “A politician based in [Hammerdal/Stockholm] posted the following on X (formerly Twitter),” and further reinforced through the politician’s user bio (“Raised and living in wonderful [Hammerdal/Stockholm]”) and a location-specific header image. Respondents indicated their level of agreement with the political statement using a 7-point Likert scale. This stage directly tests our main hypothesis (H4) concerning whether individuals are more likely to agree with policy messages when they are attributed to a politician from their geographic in-group (see Figure E1 –E2 in the Appendix for sample stimuli).
Table 1. Policy proposal statements to test place-based bias
Place-based identity was measured using a self-assessment, where respondents indicated whether they primarily viewed themselves as a “rural person” or an “urban person” (see e.g., Lunz Trujillo, Reference Lunz Trujillo2024).Footnote 1 This self-categorization guided the framing of subsequent questions. Depending on their response, participants evaluated their in-group (Stage 3: In-group identity strength), their out-group (Stage 4: Place-based resentment), and the contrast between in- and out-group (Stage 5: Place-based stereotyping)
Place-based identity strength was assessed by asking respondents a series of questions about their emotional connection to their self-identified geographic in-group, following established measures (Auerbach et al., Reference Auerbach, Eidheim and Fimreite2024a; see Table 2). Respondents who identified as rural were asked to evaluate rural people, while those who identified as urban evaluated urban people.
Table 2. Items to measure place-based identity strength, place-based resentment and place-based stereotyping
Place-based resentment was measured by asking respondents to evaluate their geographic out-group using a set of items adapted from Munis (Reference Munis2022; see Table 2). Respondents who identified more as rural answered questions about urban people, and respondents who identified themselves more as urban answered questions about rural people.
Place-based stereotyping: To gauge place-based stereotyping, respondents evaluated the extent to which participants believe people from rural and urban areas typically exhibit four personality traits: honesty, selfishness, prejudice, and intelligence (see Table 2). For each trait, respondents were asked, for example: “To what extent do you think that people from the following places are honest?” Footnote 2.Each trait is rated on a 7-point Likert scale (1 = to a very small extent, 7 = to a very large extent). Based on the answers, we constructed a Place-Based Trait Evaluation Index (PTEI). The PTEI was computed as the difference between the in- and out-group trait scores, yielding a scale from –6 to +6. A score of –6 indicates that the in-group is perceived more negatively than the out-group, while a score of +6 indicates the opposite.
At the end of the survey, we also collected demographic information, including age, gender, and education level. In addition, political orientation was measured on a 11-point continuous scale, ranging from 1 (left) to 10 (right).
Results
We present our findings in two parts: first, testing descriptive hypotheses on place-based identity, resentment, and stereotyping (H1–H3); second, evaluating the causal hypothesis on whether political place-based cues affect agreement with political statements (H4).
Descriptive results
Figure 2 presents the distribution of place-based identity strength (Figure 2A), place-based resentment scale (Figure 2B), and place-based stereotyping (Figure 2C) across rural and urban respondents. First, we hypothesized that rural respondents would exhibit stronger place-based identity than urban respondents (H1). As illustrated in Figure 2A, rural respondents reported significantly stronger in-group identification (M = 3.71, SD = 1.15) than urban respondents (M = 4.52, SD = 1.23; t(2049) = −15.32, p < 0.001, d = − 0.68). Regression results (Model 1, Table B1) showed that rural identity remained a strong predictor of place-based identity strength even after controlling for demographics and political orientation.
Second, we hypothesized that rural respondents would demonstrate higher levels of place-based resentment than urban respondents (H2). Again, we observe a similar asymmetry. Rural respondents were more likely to express higher levels of place-based resentment (M = 4.58, SD = 1.15) toward urban citizens compared to urban respondents’ resentment toward rural citizens (M = 2.76, SD = 1.11). This difference is statistically significant, t(2049) = −36.54, p < 0.001, d = − 1.62). As shown in Model 2 (Table B1), rural identity was a significant predictor of higher resentment levels, even when controlling for covariates.
Third, we hypothesized that rural respondents would show stronger trait-based stereotyping, more positive evaluations of their in-group, and more negative evaluations of their out-group than urban respondents (H3). Supporting this, rural respondents scored higher on the PTEI, with a mean of 0.75 (SD = 1.02), compared to urban respondents, who scored –0.39 (SD = 0.87). The difference is statistically significant, t(2049) = –27.23, p < .001, d = –1.20. Interestingly, the findings reveal that urban respondents evaluated their in-group less favorably than their out-group, as indicated by the negative mean on the PTEI (M = –0.39, SD = 0.87). In contrast, rural respondents showed a clear preference for their in-group over the out-group (M = 0.75, SD = 1.02), consistent with our expectations. These differences were robust across a series of regression models, reported in Table B1 (see Model 4). Together, the results highlight a significant asymmetry in group evaluations: while rural respondents exhibit strong in-group favoritism, urban respondents show a slight out-group preference.
Figure 2. A. Place-identity strength across rural and urban respondents. B. Place-based resentment across rural and urban respondents. C. Place-based stereotyping across rural and urban respondents.
The causal effect of a rural politician on policy agreement
To test H4, we examined whether agreement with political statements varied based on the geographic identity of the politician (urban vs. rural) and whether this interaction differed by the respondent’s place-based identity. Contrary to expectations, we found no evidence that respondents agreed more with statements made by in-group politicians. In fact, rural respondents agreed slightly more with statements attributed to urban politicians (M = 5.09, SD = 1.75) than with those from rural politicians (M = 4.91, SD = 1.80), though this difference was not statistically significant, t(1075) = 1.68, p = 0.094, d = 0.10. Similarly, urban respondents showed no significant difference in agreement between statements from an urban politician (M = 5.10, SD = 1.78) and a rural one (M = 4.91, SD = 1.79), t(972) = 1.67, p = .095, d = 0.11. These null effects are visualized in Figure 3. To test the robustness of these findings, we conducted regression analyses controlling for gender, age, education, and political orientation. The results confirmed that neither rural nor urban respondents expressed significantly greater agreement with politicians from their place-based in-group compared to their out-group, see Table B2 in the Appendix.
Figure 3. A. Policy agreement for rural respondents, across the urban and the rural politician. B. Policy agreement for urban respondents, across the urban and the rural politician.
Notes: Responses were measured on a 7-point Likert scale (0 = completely disagree, 7 = completely agree). Error bars show the 95% confidence intervals.
Discussion
This Registered Report has demonstrated how place-based affective polarization – operationalized through place-based identity, place-based stereotyping, and place-based resentment – manifests among rural and urban populations in Sweden. In line with previous results from Norway (Auerbach et al., Reference Auerbach, Eidheim and Fimreite2024a, Reference Auerbach, Eidheim and Fimreite2024b), findings reveal significant asymmetries between rural and urban populations, ultimately establishing three key contributions to the literature on geographic identity and political psychology.
First, we documented substantial differences in place attachment and intergroup attitudes. Rural respondents exhibit significantly higher levels of in-group identity than urban respondents, indicating that place carries greater meaning for rural citizens’ sense of self. Moreover, rural respondents demonstrate elevated levels of place-based resentment and more negative stereotypes toward urban populations compared to the reverse dynamic.
Second, we identified a crucial disconnect between affective attitudes and political behavior. Surprisingly, and contrary to our expectations, neither rural nor urban respondents showed preferential agreement with politicians sharing their geographic identity. This finding reveals that strong place-based identities and place-based resentment do not necessarily translate into corresponding place-based biases when evaluating political statements.
Third, we extend theoretical understanding of place-based identity, resentment, and stereotyping beyond established contexts. While our results align with Cramer’s (Reference Cramer2016) influential work on rural resentment in Wisconsin, we advance this research agenda by systematically examining place-based attitudes across both rural and urban populations rather than focusing solely on rural perspectives. This comparative approach strengthens the empirical foundation for understanding place-based political identities and affective polarization.
The Swedish context makes these findings particularly compelling. Sweden represents a theoretically challenging case for the study of place-based resentment, given its history of consensus-oriented governance, multi-party characteristics, and ambitious territorial equalization policies. Despite these mitigating institutional factors, we still observe robust asymmetric patterns of rural resentment. This suggests that the dynamics of place-based resentment and affective polarization originally observed in more adversarial and polarized systems like the USA are most likely relevant for considerably different political systems.
The absence of a place-based bias effect merits further consideration. Despite rural respondents’ stronger in-group identity, place-based resentment, and place-based stereotyping, which we predicted would enhance preferences for political statements from co-located politicians, we found no evidence of such a bias. This contrasts with Jacob and Munis’s (Reference Jacobs and Munis2019) findings of rural hostility toward urban-portrayed candidates in the USA. The reasons for this difference are open to debate. One possibility is institutional: the Swedish pattern may reflect the country’s institutionalized emphasis on rational, consensual, pragmatic policymaking, its multi-party characteristics, and its relatively ambitious territorial cohesion schemes, which together may contribute to dampening geographic tribalism in political evaluation. Another possibility is methodological. Our study represents, to our knowledge, the first Registered Report in this area of research. By limiting researcher flexibility in design, analysis, and reporting, Registered Reports reduces the risk of spurious findings. This raises the possibility that earlier results, while important, may in part reflect selective reporting, or post hoc analytical choices. Finally, we cannot rule out that our specific design – tying treatments to particular cities – may have activated regional rather than strictly rural/urban identities, thereby attenuating the expected effect.
Ultimately, given the support our descriptive hypotheses 1 – 3 received, our study makes place-based resentment theory somewhat more robust by demonstrating its relevance even in the least-likely cases. As Levy (2002) argues about least-likely case studies, if theories about social identity and place-based resentment succeed in Sweden’s consensual political environment, they likely have broad applicability across diverse democratic contexts. This finding expands the scope conditions for place-based resentment as a political phenomenon worthy of scholarly attention.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/XPS.2025.10021.
Data availability
Support for this research was provided by the Länsförsäkringars forskningsfond (Award no. P7_20) and Marianne & Marcus Wallenbergs Stiftelse (MMW 2019.0209). The data, code, and any additional materials required to replicate all analyses in this article are available at the Journal of Experimental Political Science Dataverse within the Harvard Dataverse Network, at: doi: 10.7910/DVN/YG4PDG and on the project’s OSF repository at https://osf.io/safgj.
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Ethics statement
We consulted the Swedish Ethical Review Authority, which determined that our research does not require ethical approval under the Swedish Ethical Review Act (2003:460) (dr number: 2024-01061-01). All methods follow relevant guidelines and regulations.
1. Use of deception and potential harms
The experimental design involved a minor form of deception. To test whether geographic cues influenced policy agreement, we randomly varied the stated origin of a political statement so that it appeared to come from a politician based in either a rural location (Hammerdal) or an urban location (Stockholm). These politicians were fictitious, and the statements were adapted to be ideologically neutral and plausible for any party or region. This manipulation was necessary to preserve internal validity and ensure that responses reflected reactions to geographic cues rather than prior knowledge of real political figures.
This form of deception is consistent with the American Political Science Association’s Principles and Guidance for Human Subjects Research, as it was minimal, posed no more than minimal risk, and could not reasonably be expected to cause harm or lasting discomfort. The potential benefits to scientific understanding of political attitudes outweighed any minimal risk from the deception.
2. Consent process and absence of debriefing
Participants were recruited via the Norstat online panel. Upon accessing the survey link, participants were presented with an information sheet describing the general purpose of the study, the procedures, the voluntary nature of participation, the estimated time commitment, confidentiality of responses, and data handling procedures. Participants indicated informed consent by actively agreeing before proceeding to the survey.
A post-survey debriefing was not conducted. The decision not to debrief participants was based on the minimal and nonsensitive nature of the deception, the low risk of harm, and the potential for a debriefing to draw unnecessary attention to the geographic cue manipulation, thereby influencing responses in any follow-up research. This approach is consistent with APSA guidance, which permits the absence of debriefing when the deception is unlikely to cause harm or distress and when disclosure could compromise the validity of ongoing or future research using similar designs.
3. Participant compensation
Participants received seven “Norstat coins” (≈ 7 SEK) for survey completion, redeemable through Norstat (the survey company) for rewards such as gift cards, cinema tickets, or charitable donations. This amount was determined by the survey company based on estimated completion time. Compensation was not contingent on responses.
4. Additional ethical considerations
The study adhered to the principles of respect for persons, beneficence, and justice as outlined in the Belmont Report. Respect for persons was ensured through informed consent and the right to withdraw without penalty. Beneficence was upheld by designing a study with minimal risk and implementing procedures to protect participant privacy and data security. Justice was promoted by recruiting a balanced sample of rural and urban residents, ensuring equitable representation. All data were collected and stored in compliance with the EU General Data Protection Regulation, and only de-identified data are publicly shared via the project’s OSF repository.
Open access
